Using Twine

Timing and scheduling

When posts go out, the random gap between them, scheduling, and what happens when a post fails.

When posts go out

Twine publishes once an hour, starting a random few minutes past the hour. A thread you post goes out at the next run, so within the hour. Twine works this way so it can sleep in between, and so posts don't land on the dot of the hour.

Gaps between posts

Every post and every reply on an account goes out after a random gap of 30 to 120 seconds. The gap is picked separately each time. A thread with four parts takes a few minutes to finish on an account, and it looks like a person posting, not a bot dump.

Each account has its own timeline. The accounts in a project post side by side, so a project with several accounts does not take longer than one account.

Schedule a thread

Without a time, a thread goes out at the next run. Schedule it for a date and time and it starts at the first run after that time, so within the hour after it. Say the time zone when you ask your agent ("tomorrow at 9:00 Madrid time").

A scheduled thread that has not started can be cancelled, in the web app or with cancel_post.

When a post fails

If a post fails because Threads or Bluesky has a problem, Twine tries it up to three times in all, a few minutes apart. If it still fails, it is marked as failed. If the access has been revoked (a deleted app password, an expired Threads login), it fails straight away and the account is marked as needing to be connected again. Either way, the rest of that thread on that account stops, because a reply cannot go out without the post before it.

Other accounts are not affected. If a thread goes to four accounts and one fails, the other three continue.

Check the status in the web app, or ask your agent (get_post) to show each account's status and link. If an account failed because it was disconnected or its access was revoked, reconnect it (see Accounts) and post again.